[
https://issues.apache.org/jira/browse/HBASE-14861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15024792#comment-15024792
]
Samir Ahmic commented on HBASE-14861:
-------------------------------------
Thank you for review [~tedyu]. I will create new patch addressing things you
mentioned.
> HBASE_ZNODE_FILE on master server is overwritten by regionserver process in
> case of master-rs collocation
> ----------------------------------------------------------------------------------------------------------
>
> Key: HBASE-14861
> URL: https://issues.apache.org/jira/browse/HBASE-14861
> Project: HBase
> Issue Type: Bug
> Components: Operability
> Affects Versions: 2.0.0
> Reporter: Samir Ahmic
> Assignee: Samir Ahmic
> Fix For: 2.0.0
>
> Attachments: HBASE-14861.patch, HBASE-14861v2.patch
>
>
> In case of master-rs collocation HBASE_ZNODE_FILE is overwritten by
> regionserver process in HRegionServer#handleReportForDutyResponse() here is
> how it looks on master server:
> {code}
> [hbase@hnode2 hbase]$ cat hbase-hbase-master.znode
> /hbase/rs/hnode2,16000,1448022074888
> {code}
> it contains regionserver znode path instead of String value of master's
> ServerName. This affects ZNodeClearer#clear() in way that will not clear
> master znode in case we detect master crash. At end this will extend
> failover time until master znode expires configured in zookeeper by
> maxSessionTimeout parameter (40s in my case).
> I have notice this on mater branch but it can be case in other branches where
> we are collocating master and rs.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
